LinuxCNC Features - a kind of NGCGUI

Problem with probe tenmplate on mill machine

Hi:

Now, Im really working with your stuff (nice!), and I've found waht I beleive it is a mistake and now I'm trying to solve (as help). The problem is the sign of the latch direction. On most of the mill machine, the Z axis is working in opposite direction. So, the conditions for latch couldn't be if the latch is bigger or lower than zero. It must be bigger than zero, and in the OPPOSITE direction to the probing direction.

(Probe down)
(Probe SUB)
(G38.2 - probe toward workpiece, stop on contact, signal error if failure)
(G38.3 - probe toward workpiece, stop on contact)
(G38.4 - probe away from workpiece, stop on loss of contact, signal error if failure)
(G38.5 - probe away from workpiece, stop on loss of contact)

O<th_probe> SUB
#<probe_func> = #1
#<feed> = #2
#<latch> = #3
#<latch_feed> = #4
(#5-#13 - axes)

G91 (incremental distance)

G#<probe_func> F#<feed> X#5 Y#6 Z#7
#<l> = [ #<latch> / SQRT[#5*#5 + #6*#6 + #7*#7] ] (probe length)

O<probe-if-00> IF [#<latch> GT 0] AND (latch towards material)
#<l> = [-#<l>]
G0 X[#5*#<l>] Y[#6*#<l>] Z[#7*#<l>]
#<l> = [-#<l>*2]
G#<probe_func> F#<latch_feed> X[#5*#<l>] Y[#6*#<l>] Z[#7*#<l>]
O<probe-if-00> ENDIF

O<probe-if-01> IF [#<latch> LT 0] (latch backwards material)
(print ["%s - %s"%/i,/ for i in [38.2,38.3,38.4,38.5]])
(Flip probe func )
#<probe_func> = [[#<probe_func>*10-380] MOD 4 / 10 + 38.2]
G#<probe_func> F#<latch_feed> X[#5*#<l>] Y[#6*#<l>] Z[#7*#<l>]
O<probe-if-01> ENDIF
G1 X[#8] F[#<_global_probe_feed>]

G90 (Absolute distance mode)

O<th_probe> ENDSUB


(O<th_probe> call [38.2] [#<_global_probe_feed>] [#<_global_latch>] [#<_global_latch_feed>] [0] [0] [-100] [5])
O<th_probe> call [38.2] [50.0 ] [-0.5] [10.0 ] [0] [0] [-100] [5]

That's is working for me:

That's is working for me:
What have you changed exactly?
As I remember if latch is less than 0 then we are latching in oposite direction. No mater of direction of the axis.
So probe in the direction
(0,0,-10) (downwards) with latch -1 will do
probing to Z-10
then probing to
Z[-10*latch] = Z[-10*-1/10] = Z1 (in relative coords, i.e. upwards)

if latch is GT 0, latch =1
probe to Z-10
go up to Z[-10 * [-latch]] = Z1
then probe to Z[2*[-latch]] = Z-2

That is how it supposed to work...
is it possible to launch directly pressing one button, or you must save the g-code and then, launche the g-code?
You will have to press refresh or select autorefresh and you'll have to open probing subroutine or remove all features and add new Probing feature.

Working perfectly!!!!
Great, can you shoot any videos?

Nick wrote:

That's is working for me:
What have you changed exactly?
I've attached the probe_file. Only one line chnage. I believe that the fix it is needed.
As I remember if latch is less than 0 then we are latching in oposite direction. No mater of direction of the axis.
So probe in the direction
(0,0,-10) (downwards) with latch -1 will do
probing to Z-10
then probing to
Z[-10*latch] = Z[-10*-1/10] = Z1 (in relative coords, i.e. upwards)

if latch is GT 0, latch =1
probe to Z-10
go up to Z[-10 * [-latch]] = Z1
then probe to Z[2*[-latch]] = Z-2

That is how it supposed to work...
is it possible to launch directly pressing one button, or you must save the g-code and then, launche the g-code?
You will have to press refresh or select autorefresh and you'll have to open probing subroutine or remove all features and add new Probing feature.

Working perfectly!!!!
Great, can you shoot any videos?

You've added: #<l> = [-#<l>]
toWhy?
Lets take probing to Z-10 and latch = -1
Then we will have
l = -1/10
and G#<probe_func> F#<latch_feed> Z[-10 * [-1/10 ]]
I.e. probe to Z1 that is seems to be fine

But if we add #<l> = [-#<l>]
we will probe on the same direction again...

Think on it.
When you are detectecting by approaching to the material, there are three moves:
- 1. Goto marterial
- 2. Go back (l = -l)
- 1. Goto material slower

When you are using the other aproaching (detecting by separating from the material):
- 1. Goto matertial
- 2. Go back slower until no contact (l must be l=-l)

When you are using the other aproaching (detecting by separating from the material):
- 1. Goto matertial
- 2. Go back slower until no contact (l must be l=-l)

Yep, but latch is already less than zerro:
IF [#<latch> LT 0]
So to latch backwards we need to probe in the direction that will point to (first probing vector)*latch, where latch is less than 0.

Nick,

I've Installed features on 2.5.3 and am testing it now. Looks really good!!! The only difference was that i had to move my features.ui file to the same directory as my linuxcnc.ini file. Is there a drawing feature for milling or is that only for lathe work? It would be nice to have the drawing feature for milling.

Great work!!!

You can try to add it into milling catalog.... but as of now I'm not sure that it will work together with multipass mill....

As of now we have to figure out what to develop next or which features should be modified...

Yeah I hear you. Looking at the source code I can try to play with what i'm thinking. Additions don't seem to be too bad.